image-generation

Name: image-generation
Availability: InStock
Author: Tolki

Generate or edit images via the Gemini 3.1 Flash Image API.

5|Updated Feb 24, 2026

Search Tags:#image-generation #text-to-image #image-to-image #gemini-api #graphic-design #branding-assets #visual-creation

Ready to run

npx skills add https://github.com/mrtolkien/GHOST --skill image-generation-mrtolkien

💬 Prompt for ChatGPT / Claude Chat▼

Please help me install this Agent Skill.
Skill: image-generation
Source: https://github.com/mrtolkien/GHOST/tree/main/prompts/skills/image-generation
Command: npx skills add https://github.com/mrtolkien/GHOST --skill image-generation-mrtolkien

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires google-genai, Pillow, and includes scripts (resource) components.

What problem does it solve?

Generate or edit images quickly using Gemini 3.1 Flash Image API, reducing manual design time and enabling rapid visual experimentation.

Core Features & Use Cases

Generate new images from prompts (text-to-image) or edit existing ones (image-to-image) with optional input references.
Supports 1K/2K/4K resolutions and automatic format handling; saves outputs to the workspace for downstream workflows.
Real-world use: create product thumbnails, marketing visuals, or concept art from concise prompts and iterate rapidly.

Quick Start

Run the generate_image.py script with a prompt to create an image, then check the saved PNG in the workspace folder.

Frequently Asked Questions about image-generation

High-intent search queries and answers about installing and using this skill.

FAQPage Schema

How do I generate images from text prompts using the Gemini API?▼

To generate images from text prompts using the Gemini API, run the generate_image.py script with your descriptive prompt. The Skill leverages the Gemini 3.1 Flash Image API to create visuals and automatically saves the output PNG file to your workspace folder.

Can I edit an existing image with Gemini 3.1 Flash Image API?▼

Yes, you can edit existing images through image-to-image processing by providing an input image alongside your prompt. This allows you to modify visuals or use reference images to guide the editing output, which is then saved to the workspace.

What resolutions are supported for text-to-image generation?▼

Text-to-image generation supports 1K, 2K, and 4K resolutions. The Skill handles automatic format processing and saves the generated visual assets in PNG format to your workspace for downstream workflows.

Do I need a Gemini API key to generate marketing visuals?▼

Yes, you need Gemini API access to generate marketing visuals or product concept art. This Skill requires a valid Gemini API connection to process prompts and input images through the Gemini 3.1 Flash Image API.

What is the best way to rapidly iterate on product concept art?▼

The best way to rapidly iterate on product concept art is using the Gemini 3.1 Flash Image API to generate visuals from concise prompts. You can quickly run the generate_image.py script repeatedly to refine designs and save iterations to your workspace.

Are there limitations when using Pillow with the Gemini API for image generation?▼

The Skill uses Pillow for image processing alongside the Gemini API, supporting 1K to 4K resolutions. Outputs are saved as PNG files in the workspace, meaning downstream workflows must read from this local directory to access the generated assets.